englishforitprofessionalsfandomcom-20200214-history
Conferences review: Participation in Theoretical and Practical Research
National Research University ”MIET” Conference “NANOELECTRONICS and NANOTECHNOLOGY” December, 16 2011 Spectroscopic Ellipsometry to measure Ti and TiN thin films. R.M.Giniatyllin ECT-65M Modern electronics have tends to reduce dimensions component of integrated circuits. They often are faced with the needs of precision measurement transparent layers thickness of complex structures and optical constants of these layers. One of the methods named the spectroscopic ellipsometry technique provides this measurement Spectroscopic ellipsometry (SE) is a very powerful tool for characterization of thin film systems. Under appropriate circumstances spectroscopic ellipsometry determines film thickness more accurately than any other known technique. In addition, spectroscopic ellipsometry measurements can provide information concerning the optical functions, surface roughness and interface layers of films. SE is also a very useful tool in measuring optical functions of bulk materials.SE measures the differences between s-polarized and p-polarized reflected light from a sample. Its strength is that at each wavelength it provides two pieces of information (namely the amplitude ratio, Ψ, and the phase difference, Δ, between the reflected components). However, the measured data from an SE experiment are usually not very interesting by themselves. The useful information, such as film thickness and optical functions, can only be determined by modeling the near-surface region of the sample and then fitting the SE data to the model using the desired parameters as variables in the numerical analysis.More materials already has final dispersion model that is used for measurement. But modern manufacturing is often faced with the needs to use new materials and creating new dispersion model for ellipsometeric measurement properly.Sometimes it is connected with fundamental research of materials properties and creative approach to solving these problems.Now we are working on the dispersion models that describe thin film titanium nitride and titanium. Structure that presents in this picture is used as diffusion barrier in new submicron technology metallization. New technology uses spectroscopic ellipsometry to control optical properties and layers thickness of this structure. But thin layers of these materials have no dispersion model and needs in creating such models. Development of supercapacitor based on carbon nanostructures. S.V.Dubkov ECT-65M'' '' e-mail: sv.dubkov@gmail.com'' '' '' Nowadays a lot of scientists try to solve the problem of storage and storing of electric energy. This is a very important matter in energetics due to nonuniformity of electric energy consumption, in car industry due to development and adoption of electromobiles anad hybrids, in home appliances and in medicine. Scientists look for new materials, create new devices for generation and energy storage, i.e. super capacitor. The main super capacitor feature is the electric capacitance that depends on electrode material as well as the electrode structure being used. Nowadays the capacity of the super capacitor is 900-3000 F. Due to the forecasts threefold increase of the capacity is possible, i.e. to use carbon nanostructures that have developed surface as the basis electrode material and to develop electrolyte or dielectric material with high permeability. During the experiments that were carried out the structures were formed using PECVD, as carbon-bearing environment СО+Н2 was used.As the result graphene petal structure was formed at 350ºС, nickel was used as catalyst. One should pay attention that the petals have high homogeneity at height 100 nm. Lowering the process temperature to 250ºС unusual carbon nanostructures in the form of posts were made from the same gas mixture. This structure has high homogeneity, posts height is 300 nm, and interval between them is 30 nm. During capacitive measurements dionized water was used as electrolyte. Steel slices were used as capacitor electrode with 5 cm2 area. On one pair of steel electrodes carbon layer in the form of posts was formed at 250ºС. Measurements showed that while using electrodes with carbon nanostructure the capacitor capacity was ~200 mkF, while using the capacitor with steel electrodes the capacitor capacity was 570 nF.Thereby, while using electrodes with developed carbon structure the capacitor capacity increased to 350 times in comparison with electrodes with undeveloped surface. In comparison with common accumulator super capacitor on the basis of carbon nanostructures has a number of advantages, i.e. less time for recharge and greater charging-discharging cycles, low cost, long life, ecological compatibility and ability to work at critical conditions. At the beginning of 2009 the world market of super capacitors was priced at about $ 300 mln. Due to different analysts’ reviews the world market annual average growth rate will be 20-25% by 2009-2015. So the market experts and experts from Agency of Industrial Information forecast the increase of the super capacitor market size up on $ 725 mln by 2015. Silicon Germanium Heterojunction Bipolar Transistor (SiGe HBT).' V.D.Yevdokimov ECT - 65M' ' SiGe technology is a perspective way to increase capacity of the analog-digital and digital-analog circuits without cost being increased a lot. Another advantage of this technology is the velocity of both bipolar and MOS transistors so that the maximum frequency could be increased. All the advantages occur because of the Ge as a dopant in some silicon regions resulting in mechanical stress in these regions and consequently in the energy zones deformation and the carriers mobility growth.The great advantage of SiGe technology is an opportunity to integrate it into conventional Si CMOS process. Silicon technology is adjusted so the yield is high. The nearest competitor of SiGe is GaAs material. It is remarkable because of high mobility of carriers in it (8500 against 1400 cm2/V∙sec in Silicon for electrons). So the velocity is high but there are many technological problems. First of all it is an absence of native oxide. Then the wafers are fragile enough. The next disadvantage is that arsenicum tends to evaporate from GaAs at high temperatures. Finally, its cost: GaAs is too expensive to become the basis of mass production. The big advantage of SiGe technology is that it could be integrated into BiCMOS process which is the main trend in the field. This technology makes it possible to do super high speed devices, which are used in the connection blocks in mobile phones: GSM module, Wi-Fi and Bluetooth. Radars, GPS and Internet modems are also using it. The bipolar transistors designed in this way are called Heterojunction Bipolar Transistors (HBT). The feature of its construction is an epitaxial base that is grown from the silicon with 10-30% germanium as a dopant. The geometrical width of the base is different for different applications. The main characteristic of any bipolar transistor is a current gain that is determined as a ratio of the collector current and base current. In modern bipolar junction transistors (BJT) current gain (beta) could achieve the value of 150-200. In HBT beta is about 400-500. So it’s evident that amplifier properties of HBT are significantly higher.The main aim of my Master’s dissertation is to optimize the structure of the SiGe HBT and analyze the results. We are going to have maximum value of the current gain and fmax. So some parameters such as Ge profile, base width etc will be varied to satisfy the requirements. X-ray fluorescent analysis E.A.Lebedev ''ECT-65'М '' X-ray spectrometry is such a thing that when a sample is irradiated with powerful primary x-rays emitted from an x-ray tube, secondary x-rays (fluorescent x-rays) will come out from this sample. The secondary x-rays are then separated into their spectral components so as to take out selectively specific fluorescent x-rays caused by the aimed element and thereby qualitative or quantitative analysis is carried out.Principle, concept, precautions and presenting procedure for BPSG/Si film thickness composition analysis using the WAFER/DISK ANALYZER 3640 are described. The ordinary applicable range of BPSG film thickness is 2500 to 9000 Å. When the optional thin film attachment is used to correct variations in B-Kα background, thinner BPSG films of 1000 to 2500 Å can be analyzed. Hereafter, the range of 2500 to 9000 Å will be called the thick BPSG film and that of 1000 to 2500 Å the thin BPSG film. Si-Kα is generated from the Si substrate and from SiO2 in the film. The X-ray intensity of Si-Kα from the film becomes higher as the film becomes thicker, but a decrease in Si-Kα intensity from the substrate due to absorption by the film is greater. As a whole, therefore, the x-ray intensity becomes lower as the film becomes thicker.The x-ray intensity of P-Kα becomes higher as the film becomes thicker.B-Kα intensity varies with film thickness when it is 3000 Å or thinner, but is almost constant when it is 3000 Å or thicker. The applicable range of the calibration curve is originally 3000 Å or thicker. But because correction is made against variations in background, the practically applicable range of the calibration curve for B2O3 concentrations (the thick BPSG film) is 2500 Å or thicker.It is known that Si-Lx is hardly generated from SiO2 (the BPSG film) and is generated from Si. Therefore, in the case of the thin BPSG film of 2500 Å or thicker, the X-ray intensity of Si-Lx from the substrate becomes higher as the film becomes thinner. '''Metallization. V.G.Plaksin –ECT-65M Today's state of the art integrated circuits contain many active and passive elements including millions of transistors, capacitors, and resistors on a single chip. These discrete elements must be connected with some form of wiring to form a circuit.Metallization is the process that connects individual devices together by means of microscopic wires to form circuits. Virtually all Integrated circuits are consist of 2-up to 6 levels of metal wiring. The ILD - interlevel dielectric layer. The ILD0 Provide dielectric insulation between silicon device components and the metal interconnect layers. The ILD 1 – dielectric layer between metal 1 and metal 2.The Contact – provides electrical connection from the interconnect to silicon components through ILD0. The Metal 1 – first layer of metal interconnect.The Via one – provides electrical connection between Metal 1 and metal 2 through ILD1.Companies commonly use the WCVD process to fill contacts/vias with tungsten. Unfortunately, if one uses WCVD to deposit tungsten directly to silicon dioxide, the tungsten flakes and peels, produce many particles. Therefore, an intermediate layer is deposited between the oxide and WCVD. Titanium reduces contact resistance. Titanium Nitride prevents tungsten from peeling. Tungsten carries current from Silicon to interconnect and called “via”.Aluminum-thin films were selected for the first 30 years of the Integrated Circuits industry. The material used in interconnects is not pure aluminum, but an aluminum alloy. Usually with Copper, sometimes with Si. Aluminum - This layer makes the contacts with the Tungsten vias. It is the primary current carrier. Titanium Nitride Layer - Creates a barrier between the Aluminum and the Titanium layers. Titanium Layer - Provides an alternate current path (shunt) around flaws in the primary current carrier. Titanium Nitride ARC Layer - This is an anti reflecting coating which aids lithography to keep control of critical dimensions and to absorb light during the resisting exposure.CONCLUSIONS. The modern metallization is multilayer multilevel complicated system. We should spend many resources, and use sophisticated methods for the forming parts of metallization such Physical Vapor Deposition, Chemical Vapor Deposition and so on. The metallization problems are very important today. Generally, these problems determinate the circuits speed. That is why we must develop this branch of the integrated circuits technology. Sources: {C}• Handbook of Multilevel Metallization for Integrated Circuits (Materials Science and Process Technology) {C}• http://www.eng.tau.ac.il/~yosish/courses.html (Professor Nathan Cheung U. C. Berkeley) 'WS 820 tool overview and the process optimization of Si3N4 wet etching process for 90 nm technology.' G.N.Potapov ECT-65M '' While examining the DNS Wet Station WS820L (Japan) we come across the possible system configurations. We can see that the station always consists of transfer unit which transports wafers through the station, loader, chemical bath, rinse bath, dryer and unloader. The following processes can be performed in the baths and dryers.The chemical bath facility includes chemical measuring bath, outer and inner process bath, circulation pump, heater and filter. The wafers are processed in the inner bath. Chemicals go through circulation pump and heater towards filter that cleans them from contamination. From the inner bath chemicals go to chemical measuring bath where their concentration, temperature and pressure of the process are set. The chemical are changed not after every process but after the period of time stipulated by the user. Partial incomplete drainage is possible as well. The next part of WS820L involves rinse bath facility, shower nozzle, N2(nitrogen gas) bubble, up-flow tube, rapid drainage valve and rinsing bath. Wafer is processed in rinsing bath where ionized water flows from shower nozzle. Approximately a minute later the water is drained through rapid drainage valve. That cycle can be repeated from 6 to 10 times. When the bath is filled with water N2 bubbles can be delivered to improve the wafers quality.Spin dryer rotates the wet wafers to dry them up with centrifugal force. It’s also possible to dry wafers with isopropyl alcohol, but this function isn’t realised on Micron.We need the configuration of bath for etching Si3N4((silicium nitride)) to form the STI isolation. The transfer unit in this station can move wafers only forward so you must configure the sequence of baths for you technology.Custom interchanging of chemical baths and rinsing baths can be possible for specified technologies. The consequence of baths for formation STI 90nm and consequence of the processes taking place there are the following. Firstly we etch SiO2(silicon oxide) (blue colour) in the bath containing HF(hydrofluoric acid), after that rinsing bath washes away chemicals and contamination from a wafer that is followed by etching in H3PO4(orthophosphoric acid) that removes Si3N4 but hardly etches SiO2. H3PO4 etches SiO3 slower than Si3N4 therefore we get STI. That is a standard process. But for the 90nm technology difference between etching speed was not enough that is why approximately 100mkm of Si3N4 is etched into H3PO4 in order to increase selectivity. The process significantly slows down etching of Si3N4 as to SiO2 (1\5 in the standard process and 1\100 in the improved process). However, that causes another problem – the solution rich with the reaction remnants leaves many particles on wafers surface that were sticking to it when the wafer was being evacuated from the bath because wafers and particles were charged unlikely during the etching process. To remove this problem bath with SC1 (PAR + DIW) is added. NH4OH characteristics help to charge the wafers and particles the same charge, so they stay in the solution after you remove wafers. 'Development of ordered nanostructures for functional electronics devices. ''' A.S.Shuliatyev ECT-65M e-mail: ashuliatyev@gmail.com Such dimensional structures, like carbon nanotubes, are very good emitters. They can be used in electrovacuum devices as source of electrons. High aspect ratio of carbon nanotubes makes the field emission properties incredible. The threshold field is reduced to 104 V/cm. They can be used in radiation-hard electronic microlamp, in gas discharge microlamp for developing a new generation of IC etc.There are some methods, which allow to get ordered nanostructures. First method is the method of fragmentation of metal nanolattices. It is used in nanoimprint lithography (NIL) to pattern metal (gold) nanogratings on a substrate. Next, a single laser pulse melts the linear structures and as a result stripes of metal disperse into array which contains round and periodic metal nanodots. As a result, the period of the nanodots in the direction of the original nanograting is determined by the period of the shallow trenches rather than the natural MIF, which leads to nanodot arrays with regular periodicity.Second method is based on porous anodic alumina. We deposit layer of aluminum on the substrate, then we anodize this layer. Second step is etching through porous anodic alumina.Our approach is as follows: NIL was made on the substrate. Then we deposit thin film of Mo. Next step was bursting photolithography. So, as a result we had a local area of Mo on the substrate. Then we deposit a thin film of Ni, which was heated.